Abstract: In view of the demand of autonomous and real time navigation system for an asteroid exploration mission, an integrated celestial autonomous navigation method for low thrust orbit maneuver is studied. Based on the analysis of engineering practice, the dynamic model with electric propulsion during orbit maneuver is established, and the autonomous navigation method combining with astronomical angle and velocity measurement is given. Aiming at overcoming the uncertainty of asteroid probe thrust, an adaptive interactive multi model unscented Kalman filter (AIMM UKF) algorithm is proposed to cover the state of the navigation system with several different models, which overcomes the influence of inaccurate prior information of the model set on the navigation accuracy and improves the robustness and anti-disturbance ability of the integrated navigation system. The performance of the algorithm is verified by mathematical simulation. The simulation results show that the presented navigation method is more precise and has less computation than the general UKF algorithm, thus demonstrating the effectiveness of the proposed method, and meeting the demand of autonomous, real time and high accuracy of the navigation system for asteroid exploration engineering missions.

Key words: Asteroid exploration, Celestial navigation, Low thrust orbit maneuver, Unscented Kalman filter, Angle and velocity measurement